This code to traverse all the gadget children elements is needed because the
old implementation will not allow a gadget's size to be reduced in Webkit. I
believe either scrollHeight or offsetHeight (in the old code path) can
always be a larger number. I think the issue here is just with the
if-condition. It should have been --
if (children[i].offsetTop != undefined && children[i].scrollHeight !=
undefined) {
On Fri, Jul 17, 2009 at 1:35 PM, Stephen Voorhees <
stephen.voorh...@autodesk.com> wrote:
> Hi All,
>
> I've been debugging an issue with dynamic-height on Safari & Chrome. On
> both of these browsers, we are seeing the height set to 0px rather than the
> gadget height on our container implementation. In debugging this I notice
> that the following method always returns 0:
>
> function getHeightForWebkit() {
> var result = 0;
> var children = document.body.childNodes;
> for (var i = 0; i < children.length; i++) {
> if (children[i].offsetTop && children[i].offsetHeight) {
> var bottom = children[i].offsetTop + children[i].scrollHeight
> + parseIntFromElemPxAttribute(children[i], "margin-bottom")
> + parseIntFromElemPxAttribute(children[i], "padding-bottom");
> result = Math.max(result, bottom);
> }
> }
> // Add margin and padding height specificed in the body (if any).
> return result
> }
>
> For our test gadget, children[i].offsetTop = 0 (offsetHeight & scrollHeight
> = 1000), and therefore regardless of the value of offsetHeight, the
> following always resolves to false:
>
> if (children[i].offsetTop && children[i].offsetHeight) {
>
> Therefore, bottom is never defined and the method always returns 0, setting
> our gadget height to 0.
>
> I'm not 100% sure how to fix it because I do not know the intent of the
> code. Should this check even be there? Should it be checking for
> 'undefined' instead? Also, why is it checking for children[i].offsetHeight
> yet then never using it?
>
> Any ideas on how to fix this would be appreciated.
>
> --Steve
>
--
mh x31813
